Instructions to use tongrow/MLX-Qwopus3.5-9B-Coder-oQ4-fp16-mtp with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tongrow/MLX-Qwopus3.5-9B-Coder-oQ4-fp16-mtp with MLX:

# Make sure mlx-vlm is installed
# pip install --upgrade mlx-vlm

from mlx_vlm import load, generate
from mlx_vlm.prompt_utils import apply_chat_template
from mlx_vlm.utils import load_config

# Load the model
model, processor = load("tongrow/MLX-Qwopus3.5-9B-Coder-oQ4-fp16-mtp")
config = load_config("tongrow/MLX-Qwopus3.5-9B-Coder-oQ4-fp16-mtp")

# Prepare input
image = ["http://images.cocodataset.org/val2017/000000039769.jpg"]
prompt = "Describe this image."

# Apply chat template
formatted_prompt = apply_chat_template(
    processor, config, prompt, num_images=1
)

# Generate output
output = generate(model, processor, formatted_prompt, image)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

How to use tongrow/MLX-Qwopus3.5-9B-Coder-oQ4-fp16-mtp with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "tongrow/MLX-Qwopus3.5-9B-Coder-oQ4-fp16-mtp"

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "tongrow/MLX-Qwopus3.5-9B-Coder-oQ4-fp16-mtp"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent

How to use tongrow/MLX-Qwopus3.5-9B-Coder-oQ4-fp16-mtp with Hermes Agent:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "tongrow/MLX-Qwopus3.5-9B-Coder-oQ4-fp16-mtp"

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default tongrow/MLX-Qwopus3.5-9B-Coder-oQ4-fp16-mtp

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use tongrow/MLX-Qwopus3.5-9B-Coder-oQ4-fp16-mtp with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "tongrow/MLX-Qwopus3.5-9B-Coder-oQ4-fp16-mtp"

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "tongrow/MLX-Qwopus3.5-9B-Coder-oQ4-fp16-mtp" \
  --custom-provider-id mlx-lm \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Performance not good from oMLX benchmark

by xht033 - opened May 24

Discussion

xht033

May 24

•

edited May 24

Intelligence Benchmark Comparison

Benchmark	Mode	Sampled	MLX-Qwopus3.5-9B-Coder-oQ4-fp16-mtp
MMLU	Sample	30/14042	20.0%
MMLU_PRO	Sample	30/12032	60.0%
TRUTHFULQA	Sample	30/817	30.0%
HUMANEVAL	Sample	30/164	40.0%

Detail

Model: MLX-Qwopus3.5-9B-Coder-oQ4-fp16-mtp

Benchmark	Accuracy	Correct	Total	Time(s)	Think
MMLU	20.0%	6	30	125.0	No
MMLU_PRO	60.0%	18	30	675.9	No
TRUTHFULQA	30.0%	9	30	83.4	No
HUMANEVAL	40.0%	12	30	889.1	No

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment